DragGAN is an AI-powered image editor that allows users to manipulate pictures by clicking and dragging
It was developed by researchers from the Max Planck Institute for Informatics, MIT, and Google
DragGAN is a point-based image manipulation tool that can incrementally move multiple points of an image along a target trajectory defined by the user
The tool uses AI to keep the output within the bounds of realism
DragGAN comprises two main components: feature-based motion supervision for handle point movement towards the target position and point tracking using specialized features to monitor the moved point.
The tool is still only a research white paper, but it has generated significant interest
Existing approaches to controlling generative adversarial networks (GANs) often lack flexibility, precision, and generality, but DragGAN offers a much less explored way of controlling GANs
DragGAN excels at precise image deformation, making complex edits as simple as drag and drop
While DragGAN is impressive, Photoshop's comprehensive tools for graphic design still hold value
DragGAN could be the next area conquered by AI in photo editing